setup.c 38.9 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
/*
 *  linux/arch/x86-64/kernel/setup.c
 *
 *  Copyright (C) 1995  Linus Torvalds
 *
 *  Nov 2001 Dave Jones <davej@suse.de>
 *  Forked from i386 setup code.
 *
 *  $Id$
 */

/*
 * This file handles the architecture-dependent parts of initialization
 */

#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/ptrace.h>
#include <linux/slab.h>
#include <linux/user.h>
#include <linux/a.out.h>
#include <linux/tty.h>
#include <linux/ioport.h>
#include <linux/delay.h>
#include <linux/config.h>
#include <linux/init.h>
#include <linux/initrd.h>
#include <linux/highmem.h>
#include <linux/bootmem.h>
#include <linux/module.h>
#include <asm/processor.h>
#include <linux/console.h>
#include <linux/seq_file.h>
38
#include <linux/crash_dump.h>
L
Linus Torvalds 已提交
39 40 41 42 43
#include <linux/root_dev.h>
#include <linux/pci.h>
#include <linux/acpi.h>
#include <linux/kallsyms.h>
#include <linux/edd.h>
44
#include <linux/mmzone.h>
45
#include <linux/kexec.h>
46
#include <linux/cpufreq.h>
47
#include <linux/dmi.h>
48
#include <linux/dma-mapping.h>
49
#include <linux/ctype.h>
50
#include <linux/suspend.h>
51

L
Linus Torvalds 已提交
52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
#include <asm/mtrr.h>
#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/io.h>
#include <asm/smp.h>
#include <asm/msr.h>
#include <asm/desc.h>
#include <video/edid.h>
#include <asm/e820.h>
#include <asm/dma.h>
#include <asm/mpspec.h>
#include <asm/mmu_context.h>
#include <asm/bootsetup.h>
#include <asm/proto.h>
#include <asm/setup.h>
#include <asm/mach_apic.h>
#include <asm/numa.h>
69
#include <asm/swiotlb.h>
70
#include <asm/sections.h>
71
#include <asm/gart-mapping.h>
72
#include <asm/dmi.h>
L
Linus Torvalds 已提交
73 74 75 76 77

/*
 * Machine setup..
 */

78
struct cpuinfo_x86 boot_cpu_data __read_mostly;
L
Linus Torvalds 已提交
79 80 81 82 83

unsigned long mmu_cr4_features;

int acpi_disabled;
EXPORT_SYMBOL(acpi_disabled);
L
Len Brown 已提交
84
#ifdef	CONFIG_ACPI
L
Linus Torvalds 已提交
85 86 87 88 89 90 91 92 93 94 95 96
extern int __initdata acpi_ht;
extern acpi_interrupt_flags	acpi_sci_flags;
int __initdata acpi_force = 0;
#endif

int acpi_numa __initdata;

/* Boot loader ID as an integer, for the benefit of proc_dointvec */
int bootloader_type;

unsigned long saved_video_mode;

97 98 99 100 101 102
/* 
 * Early DMI memory
 */
int dmi_alloc_index;
char dmi_alloc_data[DMI_MAX_DATA];

L
Linus Torvalds 已提交
103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
/*
 * Setup options
 */
struct screen_info screen_info;
struct sys_desc_table_struct {
	unsigned short length;
	unsigned char table[0];
};

struct edid_info edid_info;
struct e820map e820;

extern int root_mountflags;

char command_line[COMMAND_LINE_SIZE];

struct resource standard_io_resources[] = {
	{ .name = "dma1", .start = 0x00, .end = 0x1f,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "pic1", .start = 0x20, .end = 0x21,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "timer0", .start = 0x40, .end = 0x43,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "timer1", .start = 0x50, .end = 0x53,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "keyboard", .start = 0x60, .end = 0x6f,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "dma page reg", .start = 0x80, .end = 0x8f,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "pic2", .start = 0xa0, .end = 0xa1,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "dma2", .start = 0xc0, .end = 0xdf,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO },
	{ .name = "fpu", .start = 0xf0, .end = 0xff,
		.flags = IORESOURCE_BUSY | IORESOURCE_IO }
};

#define STANDARD_IO_RESOURCES \
	(sizeof standard_io_resources / sizeof standard_io_resources[0])

#define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)

struct resource data_resource = {
	.name = "Kernel data",
	.start = 0,
	.end = 0,
	.flags = IORESOURCE_RAM,
};
struct resource code_resource = {
	.name = "Kernel code",
	.start = 0,
	.end = 0,
	.flags = IORESOURCE_RAM,
};

#define IORESOURCE_ROM (IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM)

static struct resource system_rom_resource = {
	.name = "System ROM",
	.start = 0xf0000,
	.end = 0xfffff,
	.flags = IORESOURCE_ROM,
};

static struct resource extension_rom_resource = {
	.name = "Extension ROM",
	.start = 0xe0000,
	.end = 0xeffff,
	.flags = IORESOURCE_ROM,
};

static struct resource adapter_rom_resources[] = {
	{ .name = "Adapter ROM", .start = 0xc8000, .end = 0,
		.flags = IORESOURCE_ROM },
	{ .name = "Adapter ROM", .start = 0, .end = 0,
		.flags = IORESOURCE_ROM },
	{ .name = "Adapter ROM", .start = 0, .end = 0,
		.flags = IORESOURCE_ROM },
	{ .name = "Adapter ROM", .start = 0, .end = 0,
		.flags = IORESOURCE_ROM },
	{ .name = "Adapter ROM", .start = 0, .end = 0,
		.flags = IORESOURCE_ROM },
	{ .name = "Adapter ROM", .start = 0, .end = 0,
		.flags = IORESOURCE_ROM }
};

#define ADAPTER_ROM_RESOURCES \
	(sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])

static struct resource video_rom_resource = {
	.name = "Video ROM",
	.start = 0xc0000,
	.end = 0xc7fff,
	.flags = IORESOURCE_ROM,
};

static struct resource video_ram_resource = {
	.name = "Video RAM area",
	.start = 0xa0000,
	.end = 0xbffff,
	.flags = IORESOURCE_RAM,
};

#define romsignature(x) (*(unsigned short *)(x) == 0xaa55)

static int __init romchecksum(unsigned char *rom, unsigned long length)
{
	unsigned char *p, sum = 0;

	for (p = rom; p < rom + length; p++)
		sum += *p;
	return sum == 0;
}

static void __init probe_roms(void)
{
	unsigned long start, length, upper;
	unsigned char *rom;
	int	      i;

	/* video rom */
	upper = adapter_rom_resources[0].start;
	for (start = video_rom_resource.start; start < upper; start += 2048) {
		rom = isa_bus_to_virt(start);
		if (!romsignature(rom))
			continue;

		video_rom_resource.start = start;

		/* 0 < length <= 0x7f * 512, historically */
		length = rom[2] * 512;

		/* if checksum okay, trust length byte */
		if (length && romchecksum(rom, length))
			video_rom_resource.end = start + length - 1;

		request_resource(&iomem_resource, &video_rom_resource);
		break;
			}

	start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
	if (start < upper)
		start = upper;

	/* system rom */
	request_resource(&iomem_resource, &system_rom_resource);
	upper = system_rom_resource.start;

	/* check for extension rom (ignore length byte!) */
	rom = isa_bus_to_virt(extension_rom_resource.start);
	if (romsignature(rom)) {
		length = extension_rom_resource.end - extension_rom_resource.start + 1;
		if (romchecksum(rom, length)) {
			request_resource(&iomem_resource, &extension_rom_resource);
			upper = extension_rom_resource.start;
		}
	}

	/* check for adapter roms on 2k boundaries */
	for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
		rom = isa_bus_to_virt(start);
		if (!romsignature(rom))
			continue;

		/* 0 < length <= 0x7f * 512, historically */
		length = rom[2] * 512;

		/* but accept any length that fits if checksum okay */
		if (!length || start + length > upper || !romchecksum(rom, length))
			continue;

		adapter_rom_resources[i].start = start;
		adapter_rom_resources[i].end = start + length - 1;
		request_resource(&iomem_resource, &adapter_rom_resources[i]);

		start = adapter_rom_resources[i++].end & ~2047UL;
	}
}

282 283 284 285 286 287 288
/* Check for full argument with no trailing characters */
static int fullarg(char *p, char *arg)
{
	int l = strlen(arg);
	return !memcmp(p, arg, l) && (p[l] == 0 || isspace(p[l]));
}

L
Linus Torvalds 已提交
289 290 291 292
static __init void parse_cmdline_early (char ** cmdline_p)
{
	char c = ' ', *to = command_line, *from = COMMAND_LINE;
	int len = 0;
293
	int userdef = 0;
L
Linus Torvalds 已提交
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309

	for (;;) {
		if (c != ' ') 
			goto next_char; 

#ifdef  CONFIG_SMP
		/*
		 * If the BIOS enumerates physical processors before logical,
		 * maxcpus=N at enumeration-time can be used to disable HT.
		 */
		else if (!memcmp(from, "maxcpus=", 8)) {
			extern unsigned int maxcpus;

			maxcpus = simple_strtoul(from + 8, NULL, 0);
		}
#endif
L
Len Brown 已提交
310
#ifdef CONFIG_ACPI
L
Linus Torvalds 已提交
311
		/* "acpi=off" disables both ACPI table parsing and interpreter init */
312
		if (fullarg(from,"acpi=off"))
L
Linus Torvalds 已提交
313 314
			disable_acpi();

315
		if (fullarg(from, "acpi=force")) { 
L
Linus Torvalds 已提交
316 317 318 319 320 321 322
			/* add later when we do DMI horrors: */
			acpi_force = 1;
			acpi_disabled = 0;
		}

		/* acpi=ht just means: do ACPI MADT parsing 
		   at bootup, but don't enable the full ACPI interpreter */
323
		if (fullarg(from, "acpi=ht")) { 
L
Linus Torvalds 已提交
324 325 326 327
			if (!acpi_force)
				disable_acpi();
			acpi_ht = 1; 
		}
328
                else if (fullarg(from, "pci=noacpi")) 
L
Linus Torvalds 已提交
329
			acpi_disable_pci();
330
		else if (fullarg(from, "acpi=noirq"))
L
Linus Torvalds 已提交
331 332
			acpi_noirq_set();

333
		else if (fullarg(from, "acpi_sci=edge"))
L
Linus Torvalds 已提交
334
			acpi_sci_flags.trigger =  1;
335
		else if (fullarg(from, "acpi_sci=level"))
L
Linus Torvalds 已提交
336
			acpi_sci_flags.trigger = 3;
337
		else if (fullarg(from, "acpi_sci=high"))
L
Linus Torvalds 已提交
338
			acpi_sci_flags.polarity = 1;
339
		else if (fullarg(from, "acpi_sci=low"))
L
Linus Torvalds 已提交
340 341 342
			acpi_sci_flags.polarity = 3;

		/* acpi=strict disables out-of-spec workarounds */
343
		else if (fullarg(from, "acpi=strict")) {
L
Linus Torvalds 已提交
344 345
			acpi_strict = 1;
		}
346
#ifdef CONFIG_X86_IO_APIC
347
		else if (fullarg(from, "acpi_skip_timer_override"))
348 349
			acpi_skip_timer_override = 1;
#endif
L
Linus Torvalds 已提交
350 351
#endif

352
		if (fullarg(from, "disable_timer_pin_1"))
353
			disable_timer_pin_1 = 1;
354
		if (fullarg(from, "enable_timer_pin_1"))
355 356
			disable_timer_pin_1 = -1;

357 358
		if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
			clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
L
Linus Torvalds 已提交
359
			disable_apic = 1;
360
		}
L
Linus Torvalds 已提交
361

362
		if (fullarg(from, "noapic"))
L
Linus Torvalds 已提交
363 364
			skip_ioapic_setup = 1;

365
		if (fullarg(from,"apic")) {
L
Linus Torvalds 已提交
366 367 368 369 370 371 372
			skip_ioapic_setup = 0;
			ioapic_force = 1;
		}
			
		if (!memcmp(from, "mem=", 4))
			parse_memopt(from+4, &from); 

373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
		if (!memcmp(from, "memmap=", 7)) {
			/* exactmap option is for used defined memory */
			if (!memcmp(from+7, "exactmap", 8)) {
#ifdef CONFIG_CRASH_DUMP
				/* If we are doing a crash dump, we
				 * still need to know the real mem
				 * size before original memory map is
				 * reset.
				 */
				saved_max_pfn = e820_end_of_ram();
#endif
				from += 8+7;
				end_pfn_map = 0;
				e820.nr_map = 0;
				userdef = 1;
			}
			else {
				parse_memmapopt(from+7, &from);
				userdef = 1;
			}
		}

395
#ifdef CONFIG_NUMA
L
Linus Torvalds 已提交
396 397 398 399 400 401 402 403
		if (!memcmp(from, "numa=", 5))
			numa_setup(from+5); 
#endif

		if (!memcmp(from,"iommu=",6)) { 
			iommu_setup(from+6); 
		}

404
		if (fullarg(from,"oops=panic"))
L
Linus Torvalds 已提交
405 406 407 408 409
			panic_on_oops = 1;

		if (!memcmp(from, "noexec=", 7))
			nonx_setup(from + 7);

410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
#ifdef CONFIG_KEXEC
		/* crashkernel=size@addr specifies the location to reserve for
		 * a crash kernel.  By reserving this memory we guarantee
		 * that linux never set's it up as a DMA target.
		 * Useful for holding code to do something appropriate
		 * after a kernel panic.
		 */
		else if (!memcmp(from, "crashkernel=", 12)) {
			unsigned long size, base;
			size = memparse(from+12, &from);
			if (*from == '@') {
				base = memparse(from+1, &from);
				/* FIXME: Do I want a sanity check
				 * to validate the memory range?
				 */
				crashk_res.start = base;
				crashk_res.end   = base + size - 1;
			}
		}
#endif

431 432 433 434 435 436 437 438
#ifdef CONFIG_PROC_VMCORE
		/* elfcorehdr= specifies the location of elf core header
		 * stored by the crashed kernel. This option will be passed
		 * by kexec loader to the capture kernel.
		 */
		else if(!memcmp(from, "elfcorehdr=", 11))
			elfcorehdr_addr = memparse(from+11, &from);
#endif
439

440
#ifdef CONFIG_HOTPLUG_CPU
441 442 443 444
		else if (!memcmp(from, "additional_cpus=", 16))
			setup_additional_cpus(from+16);
#endif

L
Linus Torvalds 已提交
445 446 447 448 449 450 451 452
	next_char:
		c = *(from++);
		if (!c)
			break;
		if (COMMAND_LINE_SIZE <= ++len)
			break;
		*(to++) = c;
	}
453 454 455 456
	if (userdef) {
		printk(KERN_INFO "user-defined physical RAM map:\n");
		e820_print_map("user");
	}
L
Linus Torvalds 已提交
457 458 459 460
	*to = '\0';
	*cmdline_p = command_line;
}

461
#ifndef CONFIG_NUMA
462 463
static void __init
contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
L
Linus Torvalds 已提交
464
{
465 466 467 468 469 470 471 472 473
	unsigned long bootmap_size, bootmap;

	bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
	bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size);
	if (bootmap == -1L)
		panic("Cannot find bootmem map of size %ld\n",bootmap_size);
	bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
	e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
	reserve_bootmem(bootmap, bootmap_size);
L
Linus Torvalds 已提交
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
} 
#endif

/* Use inline assembly to define this because the nops are defined 
   as inline assembly strings in the include files and we cannot 
   get them easily into strings. */
asm("\t.data\nk8nops: " 
    K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
    K8_NOP7 K8_NOP8); 
    
extern unsigned char k8nops[];
static unsigned char *k8_nops[ASM_NOP_MAX+1] = { 
     NULL,
     k8nops,
     k8nops + 1,
     k8nops + 1 + 2,
     k8nops + 1 + 2 + 3,
     k8nops + 1 + 2 + 3 + 4,
     k8nops + 1 + 2 + 3 + 4 + 5,
     k8nops + 1 + 2 + 3 + 4 + 5 + 6,
     k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
}; 

497 498
extern char __vsyscall_0;

L
Linus Torvalds 已提交
499 500 501 502 503 504 505 506 507 508 509
/* Replace instructions with better alternatives for this CPU type.

   This runs before SMP is initialized to avoid SMP problems with
   self modifying code. This implies that assymetric systems where
   APs have less capabilities than the boot processor are not handled. 
   In this case boot with "noreplacement". */ 
void apply_alternatives(void *start, void *end) 
{ 
	struct alt_instr *a; 
	int diff, i, k;
	for (a = start; (void *)a < end; a++) { 
510 511
		u8 *instr;

L
Linus Torvalds 已提交
512 513 514 515
		if (!boot_cpu_has(a->cpuid))
			continue;

		BUG_ON(a->replacementlen > a->instrlen); 
516 517 518 519 520
		instr = a->instr;
		/* vsyscall code is not mapped yet. resolve it manually. */
		if (instr >= (u8 *)VSYSCALL_START && instr < (u8*)VSYSCALL_END)
			instr = __va(instr - (u8*)VSYSCALL_START + (u8*)__pa_symbol(&__vsyscall_0));
		__inline_memcpy(instr, a->replacement, a->replacementlen);
L
Linus Torvalds 已提交
521 522 523 524 525 526 527
		diff = a->instrlen - a->replacementlen; 

		/* Pad the rest with nops */
		for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
			k = diff;
			if (k > ASM_NOP_MAX)
				k = ASM_NOP_MAX;
528
			__inline_memcpy(instr + i, k8_nops[k], k);
L
Linus Torvalds 已提交
529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
		} 
	}
} 

static int no_replacement __initdata = 0; 
 
void __init alternative_instructions(void)
{
	extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
	if (no_replacement) 
		return;
	apply_alternatives(__alt_instructions, __alt_instructions_end);
}

static int __init noreplacement_setup(char *s)
{ 
     no_replacement = 1; 
546
     return 1;
L
Linus Torvalds 已提交
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
} 

__setup("noreplacement", noreplacement_setup); 

#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
struct edd edd;
#ifdef CONFIG_EDD_MODULE
EXPORT_SYMBOL(edd);
#endif
/**
 * copy_edd() - Copy the BIOS EDD information
 *              from boot_params into a safe place.
 *
 */
static inline void copy_edd(void)
{
     memcpy(edd.mbr_signature, EDD_MBR_SIGNATURE, sizeof(edd.mbr_signature));
     memcpy(edd.edd_info, EDD_BUF, sizeof(edd.edd_info));
     edd.mbr_signature_nr = EDD_MBR_SIG_NR;
     edd.edd_info_nr = EDD_NR;
}
#else
static inline void copy_edd(void)
{
}
#endif

#define EBDA_ADDR_POINTER 0x40E
575 576 577 578 579

unsigned __initdata ebda_addr;
unsigned __initdata ebda_size;

static void discover_ebda(void)
L
Linus Torvalds 已提交
580
{
581
	/*
L
Linus Torvalds 已提交
582 583 584
	 * there is a real-mode segmented pointer pointing to the 
	 * 4K EBDA area at 0x40E
	 */
585 586 587 588 589 590 591 592 593 594 595 596
	ebda_addr = *(unsigned short *)EBDA_ADDR_POINTER;
	ebda_addr <<= 4;

	ebda_size = *(unsigned short *)(unsigned long)ebda_addr;

	/* Round EBDA up to pages */
	if (ebda_size == 0)
		ebda_size = 1;
	ebda_size <<= 10;
	ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
	if (ebda_size > 64*1024)
		ebda_size = 64*1024;
L
Linus Torvalds 已提交
597 598
}

599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692
#ifdef CONFIG_SOFTWARE_SUSPEND
static void __init mark_nosave_page_range(unsigned long start, unsigned long end)
{
	struct page *page;
	while (start <= end) {
		page = pfn_to_page(start);
		SetPageNosave(page);
		start++;
	}
}

static void __init e820_nosave_reserved_pages(void)
{
	int i;
	unsigned long r_start = 0, r_end = 0;

	/* Assume e820 map is sorted */
	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
		unsigned long start, end;

		start = round_down(ei->addr, PAGE_SIZE);
		end = round_up(ei->addr + ei->size, PAGE_SIZE);
		if (start >= end)
			continue;
		if (ei->type == E820_RESERVED)
			continue;
		r_end = start>>PAGE_SHIFT;
		/* swsusp ignores invalid pfn, ignore these pages here */
		if (r_end > end_pfn)
			r_end = end_pfn;
		if (r_end > r_start)
			mark_nosave_page_range(r_start, r_end-1);
		if (r_end >= end_pfn)
			break;
		r_start = end>>PAGE_SHIFT;
	}
}

static void __init e820_save_acpi_pages(void)
{
	int i;

	/* Assume e820 map is sorted */
	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
		unsigned long start, end;

		start = ei->addr, PAGE_SIZE;
		end = ei->addr + ei->size;
		if (start >= end)
			continue;
		if (ei->type != E820_ACPI && ei->type != E820_NVS)
			continue;
		/*
		 * If the region is below end_pfn, it will be
		 * saved/restored by swsusp follow 'RAM' type.
		 */
		if (start < (end_pfn << PAGE_SHIFT))
			start = end_pfn << PAGE_SHIFT;
		if (end > start)
			swsusp_add_arch_pages(start, end);
	}
}

extern char __start_rodata, __end_rodata;
/*
 * BIOS reserved region/hole - no save/restore
 * ACPI NVS - save/restore
 * ACPI Data - this is a little tricky, the mem could be used by OS after OS
 * reads tables from the region, but anyway save/restore the memory hasn't any
 * side effect and Linux runtime module load/unload might use it.
 * kernel rodata - no save/restore (kernel rodata isn't changed)
 */
static int __init mark_nosave_pages(void)
{
	unsigned long pfn_start, pfn_end;

	/* BIOS reserved regions & holes */
	e820_nosave_reserved_pages();

	/* kernel rodata */
	pfn_start = round_up(__pa_symbol(&__start_rodata), PAGE_SIZE) >> PAGE_SHIFT;
	pfn_end = round_down(__pa_symbol(&__end_rodata), PAGE_SIZE) >> PAGE_SHIFT;
	mark_nosave_page_range(pfn_start, pfn_end-1);

	/* record ACPI Data/NVS as saveable */
	e820_save_acpi_pages();

	return 0;
}
core_initcall(mark_nosave_pages);
#endif

L
Linus Torvalds 已提交
693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
void __init setup_arch(char **cmdline_p)
{
	unsigned long kernel_end;

 	ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
 	screen_info = SCREEN_INFO;
	edid_info = EDID_INFO;
	saved_video_mode = SAVED_VIDEO_MODE;
	bootloader_type = LOADER_TYPE;

#ifdef CONFIG_BLK_DEV_RAM
	rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
	rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
	rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
#endif
	setup_memory_region();
	copy_edd();

	if (!MOUNT_ROOT_RDONLY)
		root_mountflags &= ~MS_RDONLY;
	init_mm.start_code = (unsigned long) &_text;
	init_mm.end_code = (unsigned long) &_etext;
	init_mm.end_data = (unsigned long) &_edata;
	init_mm.brk = (unsigned long) &_end;

	code_resource.start = virt_to_phys(&_text);
	code_resource.end = virt_to_phys(&_etext)-1;
	data_resource.start = virt_to_phys(&_etext);
	data_resource.end = virt_to_phys(&_edata)-1;

	parse_cmdline_early(cmdline_p);

	early_identify_cpu(&boot_cpu_data);

	/*
	 * partially used pages are not usable - thus
	 * we are rounding upwards:
	 */
	end_pfn = e820_end_of_ram();
732
	num_physpages = end_pfn;		/* for pfn_valid */
L
Linus Torvalds 已提交
733 734 735

	check_efer();

736 737
	discover_ebda();

L
Linus Torvalds 已提交
738 739
	init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));

740 741
	dmi_scan_machine();

742 743
	zap_low_mappings(0);

L
Len Brown 已提交
744
#ifdef CONFIG_ACPI
L
Linus Torvalds 已提交
745 746 747 748 749 750 751 752 753 754 755 756 757 758
	/*
	 * Initialize the ACPI boot-time table parser (gets the RSDP and SDT).
	 * Call this early for SRAT node setup.
	 */
	acpi_boot_table_init();
#endif

#ifdef CONFIG_ACPI_NUMA
	/*
	 * Parse SRAT to discover nodes.
	 */
	acpi_numa_init();
#endif

759
#ifdef CONFIG_NUMA
L
Linus Torvalds 已提交
760 761
	numa_initmem_init(0, end_pfn); 
#else
762
	contig_initmem_init(0, end_pfn);
L
Linus Torvalds 已提交
763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779
#endif

	/* Reserve direct mapping */
	reserve_bootmem_generic(table_start << PAGE_SHIFT, 
				(table_end - table_start) << PAGE_SHIFT);

	/* reserve kernel */
	kernel_end = round_up(__pa_symbol(&_end),PAGE_SIZE);
	reserve_bootmem_generic(HIGH_MEMORY, kernel_end - HIGH_MEMORY);

	/*
	 * reserve physical page 0 - it's a special BIOS page on many boxes,
	 * enabling clean reboots, SMP operation, laptop functions.
	 */
	reserve_bootmem_generic(0, PAGE_SIZE);

	/* reserve ebda region */
780 781
	if (ebda_addr)
		reserve_bootmem_generic(ebda_addr, ebda_size);
L
Linus Torvalds 已提交
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823

#ifdef CONFIG_SMP
	/*
	 * But first pinch a few for the stack/trampoline stuff
	 * FIXME: Don't need the extra page at 4K, but need to fix
	 * trampoline before removing it. (see the GDT stuff)
	 */
	reserve_bootmem_generic(PAGE_SIZE, PAGE_SIZE);

	/* Reserve SMP trampoline */
	reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, PAGE_SIZE);
#endif

#ifdef CONFIG_ACPI_SLEEP
       /*
        * Reserve low memory region for sleep support.
        */
       acpi_reserve_bootmem();
#endif
#ifdef CONFIG_X86_LOCAL_APIC
	/*
	 * Find and reserve possible boot-time SMP configuration:
	 */
	find_smp_config();
#endif
#ifdef CONFIG_BLK_DEV_INITRD
	if (LOADER_TYPE && INITRD_START) {
		if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
			reserve_bootmem_generic(INITRD_START, INITRD_SIZE);
			initrd_start =
				INITRD_START ? INITRD_START + PAGE_OFFSET : 0;
			initrd_end = initrd_start+INITRD_SIZE;
		}
		else {
			printk(KERN_ERR "initrd extends beyond end of memory "
			    "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
			    (unsigned long)(INITRD_START + INITRD_SIZE),
			    (unsigned long)(end_pfn << PAGE_SHIFT));
			initrd_start = 0;
		}
	}
#endif
824 825
#ifdef CONFIG_KEXEC
	if (crashk_res.start != crashk_res.end) {
826
		reserve_bootmem_generic(crashk_res.start,
827 828 829
			crashk_res.end - crashk_res.start + 1);
	}
#endif
830

L
Linus Torvalds 已提交
831 832 833 834
	paging_init();

	check_ioapic();

835 836 837 838 839 840
	/*
	 * set this early, so we dont allocate cpu0
	 * if MADT list doesnt list BSP first
	 * mpparse.c/MP_processor_info() allocates logical cpu numbers.
	 */
	cpu_set(0, cpu_present_map);
L
Len Brown 已提交
841
#ifdef CONFIG_ACPI
L
Linus Torvalds 已提交
842 843 844 845 846 847
	/*
	 * Read APIC and some other early information from ACPI tables.
	 */
	acpi_boot_init();
#endif

848 849
	init_cpu_to_node();

L
Linus Torvalds 已提交
850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874
#ifdef CONFIG_X86_LOCAL_APIC
	/*
	 * get boot-time SMP configuration:
	 */
	if (smp_found_config)
		get_smp_config();
	init_apic_mappings();
#endif

	/*
	 * Request address space for all standard RAM and ROM resources
	 * and also for regions reported as reserved by the e820.
	 */
	probe_roms();
	e820_reserve_resources(); 

	request_resource(&iomem_resource, &video_ram_resource);

	{
	unsigned i;
	/* request I/O space for devices used on all i[345]86 PCs */
	for (i = 0; i < STANDARD_IO_RESOURCES; i++)
		request_resource(&ioport_resource, &standard_io_resources[i]);
	}

875
	e820_setup_gap();
L
Linus Torvalds 已提交
876 877

#ifdef CONFIG_GART_IOMMU
J
Jon Mason 已提交
878
	iommu_hole_init();
L
Linus Torvalds 已提交
879 880 881 882 883 884 885 886 887 888 889
#endif

#ifdef CONFIG_VT
#if defined(CONFIG_VGA_CONSOLE)
	conswitchp = &vga_con;
#elif defined(CONFIG_DUMMY_CONSOLE)
	conswitchp = &dummy_con;
#endif
#endif
}

890
static int __cpuinit get_model_name(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
891 892 893
{
	unsigned int *v;

894
	if (c->extended_cpuid_level < 0x80000004)
L
Linus Torvalds 已提交
895 896 897 898 899 900 901 902 903 904 905
		return 0;

	v = (unsigned int *) c->x86_model_id;
	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
	cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
	cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
	c->x86_model_id[48] = 0;
	return 1;
}


906
static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
907 908 909
{
	unsigned int n, dummy, eax, ebx, ecx, edx;

910
	n = c->extended_cpuid_level;
L
Linus Torvalds 已提交
911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939

	if (n >= 0x80000005) {
		cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
			edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
		c->x86_cache_size=(ecx>>24)+(edx>>24);
		/* On K8 L1 TLB is inclusive, so don't count it */
		c->x86_tlbsize = 0;
	}

	if (n >= 0x80000006) {
		cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
		ecx = cpuid_ecx(0x80000006);
		c->x86_cache_size = ecx >> 16;
		c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);

		printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
		c->x86_cache_size, ecx & 0xFF);
	}

	if (n >= 0x80000007)
		cpuid(0x80000007, &dummy, &dummy, &dummy, &c->x86_power); 
	if (n >= 0x80000008) {
		cpuid(0x80000008, &eax, &dummy, &dummy, &dummy); 
		c->x86_virt_bits = (eax >> 8) & 0xff;
		c->x86_phys_bits = eax & 0xff;
	}
}

940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957
#ifdef CONFIG_NUMA
static int nearby_node(int apicid)
{
	int i;
	for (i = apicid - 1; i >= 0; i--) {
		int node = apicid_to_node[i];
		if (node != NUMA_NO_NODE && node_online(node))
			return node;
	}
	for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
		int node = apicid_to_node[i];
		if (node != NUMA_NO_NODE && node_online(node))
			return node;
	}
	return first_node(node_online_map); /* Shouldn't happen */
}
#endif

958 959 960 961 962 963 964
/*
 * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
 * Assumes number of cores is a power of two.
 */
static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
{
#ifdef CONFIG_SMP
965
	int cpu = smp_processor_id();
966
	unsigned bits;
967 968
#ifdef CONFIG_NUMA
	int node = 0;
969
	unsigned apicid = hard_smp_processor_id();
970
#endif
971 972

	bits = 0;
973
	while ((1 << bits) < c->x86_max_cores)
974 975 976 977 978
		bits++;

	/* Low order bits define the core id (index of core in socket) */
	cpu_core_id[cpu] = phys_proc_id[cpu] & ((1 << bits)-1);
	/* Convert the APIC ID into the socket ID */
979
	phys_proc_id[cpu] = phys_pkg_id(bits);
980 981

#ifdef CONFIG_NUMA
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002
  	node = phys_proc_id[cpu];
 	if (apicid_to_node[apicid] != NUMA_NO_NODE)
 		node = apicid_to_node[apicid];
 	if (!node_online(node)) {
 		/* Two possibilities here:
 		   - The CPU is missing memory and no node was created.
 		   In that case try picking one from a nearby CPU
 		   - The APIC IDs differ from the HyperTransport node IDs
 		   which the K8 northbridge parsing fills in.
 		   Assume they are all increased by a constant offset,
 		   but in the same order as the HT nodeids.
 		   If that doesn't result in a usable node fall back to the
 		   path for the previous case.  */
 		int ht_nodeid = apicid - (phys_proc_id[0] << bits);
 		if (ht_nodeid >= 0 &&
 		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
 			node = apicid_to_node[ht_nodeid];
 		/* Pick a nearby node */
 		if (!node_online(node))
 			node = nearby_node(apicid);
 	}
1003
	numa_set_node(cpu, node);
1004

1005 1006
  	printk(KERN_INFO "CPU %d/%x(%d) -> Node %d -> Core %d\n",
  			cpu, apicid, c->x86_max_cores, node, cpu_core_id[cpu]);
1007 1008 1009
#endif
#endif
}
L
Linus Torvalds 已提交
1010 1011 1012 1013

static int __init init_amd(struct cpuinfo_x86 *c)
{
	int r;
1014
	unsigned level;
L
Linus Torvalds 已提交
1015

1016 1017 1018
#ifdef CONFIG_SMP
	unsigned long value;

1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
	/*
	 * Disable TLB flush filter by setting HWCR.FFDIS on K8
	 * bit 6 of msr C001_0015
 	 *
	 * Errata 63 for SH-B3 steppings
	 * Errata 122 for all steppings (F+ have it disabled by default)
	 */
	if (c->x86 == 15) {
		rdmsrl(MSR_K8_HWCR, value);
		value |= 1 << 6;
		wrmsrl(MSR_K8_HWCR, value);
	}
1031 1032
#endif

L
Linus Torvalds 已提交
1033 1034 1035 1036
	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
	clear_bit(0*32+31, &c->x86_capability);
	
1037 1038 1039 1040 1041
	/* On C+ stepping K8 rep microcode works well for copy/memset */
	level = cpuid_eax(1);
	if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58))
		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);

1042 1043 1044 1045
	/* Enable workaround for FXSAVE leak */
	if (c->x86 >= 6)
		set_bit(X86_FEATURE_FXSAVE_LEAK, &c->x86_capability);

L
Linus Torvalds 已提交
1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
	r = get_model_name(c);
	if (!r) { 
		switch (c->x86) { 
		case 15:
			/* Should distinguish Models here, but this is only
			   a fallback anyways. */
			strcpy(c->x86_model_id, "Hammer");
			break; 
		} 
	} 
	display_cacheinfo(c);

1058 1059 1060 1061
	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
	if (c->x86_power & (1<<8))
		set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);

1062
	if (c->extended_cpuid_level >= 0x80000008) {
1063
		c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
L
Linus Torvalds 已提交
1064

1065
		amd_detect_cmp(c);
L
Linus Torvalds 已提交
1066 1067 1068 1069 1070
	}

	return r;
}

1071
static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1072 1073 1074
{
#ifdef CONFIG_SMP
	u32 	eax, ebx, ecx, edx;
1075
	int 	index_msb, core_bits;
L
Linus Torvalds 已提交
1076
	int 	cpu = smp_processor_id();
1077 1078 1079 1080

	cpuid(1, &eax, &ebx, &ecx, &edx);


1081
	if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
L
Linus Torvalds 已提交
1082 1083 1084
		return;

	smp_num_siblings = (ebx & 0xff0000) >> 16;
1085

L
Linus Torvalds 已提交
1086 1087
	if (smp_num_siblings == 1) {
		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
1088 1089
	} else if (smp_num_siblings > 1 ) {

L
Linus Torvalds 已提交
1090 1091 1092 1093 1094
		if (smp_num_siblings > NR_CPUS) {
			printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings);
			smp_num_siblings = 1;
			return;
		}
1095 1096

		index_msb = get_count_order(smp_num_siblings);
L
Linus Torvalds 已提交
1097
		phys_proc_id[cpu] = phys_pkg_id(index_msb);
1098

L
Linus Torvalds 已提交
1099 1100
		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
		       phys_proc_id[cpu]);
1101

1102
		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
1103

1104 1105 1106
		index_msb = get_count_order(smp_num_siblings) ;

		core_bits = get_count_order(c->x86_max_cores);
1107

1108 1109
		cpu_core_id[cpu] = phys_pkg_id(index_msb) &
					       ((1 << core_bits) - 1);
1110

1111
		if (c->x86_max_cores > 1)
1112 1113
			printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
			       cpu_core_id[cpu]);
L
Linus Torvalds 已提交
1114 1115 1116 1117
	}
#endif
}

1118 1119 1120
/*
 * find out the number of processor cores on the die
 */
1121
static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
{
	unsigned int eax;

	if (c->cpuid_level < 4)
		return 1;

	__asm__("cpuid"
		: "=a" (eax)
		: "0" (4), "c" (0)
		: "bx", "dx");

	if (eax & 0x1f)
		return ((eax >> 26) + 1);
	else
		return 1;
}

1139 1140 1141
static void srat_detect_node(void)
{
#ifdef CONFIG_NUMA
1142
	unsigned node;
1143 1144 1145 1146
	int cpu = smp_processor_id();

	/* Don't do the funky fallback heuristics the AMD version employs
	   for now. */
1147
	node = apicid_to_node[hard_smp_processor_id()];
1148
	if (node == NUMA_NO_NODE)
1149
		node = first_node(node_online_map);
1150
	numa_set_node(cpu, node);
1151 1152 1153 1154 1155 1156

	if (acpi_numa > 0)
		printk(KERN_INFO "CPU %d -> Node %d\n", cpu, node);
#endif
}

1157
static void __cpuinit init_intel(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1158 1159 1160 1161 1162
{
	/* Cache sizes */
	unsigned n;

	init_intel_cacheinfo(c);
1163
	n = c->extended_cpuid_level;
L
Linus Torvalds 已提交
1164 1165 1166 1167
	if (n >= 0x80000008) {
		unsigned eax = cpuid_eax(0x80000008);
		c->x86_virt_bits = (eax >> 8) & 0xff;
		c->x86_phys_bits = eax & 0xff;
1168 1169 1170 1171 1172
		/* CPUID workaround for Intel 0F34 CPU */
		if (c->x86_vendor == X86_VENDOR_INTEL &&
		    c->x86 == 0xF && c->x86_model == 0x3 &&
		    c->x86_mask == 0x4)
			c->x86_phys_bits = 36;
L
Linus Torvalds 已提交
1173 1174 1175 1176
	}

	if (c->x86 == 15)
		c->x86_cache_alignment = c->x86_clflush_size * 2;
1177 1178
	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
1179
		set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
1180
	set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
1181
 	c->x86_max_cores = intel_num_cpu_cores(c);
1182 1183

	srat_detect_node();
L
Linus Torvalds 已提交
1184 1185
}

1186
static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206
{
	char *v = c->x86_vendor_id;

	if (!strcmp(v, "AuthenticAMD"))
		c->x86_vendor = X86_VENDOR_AMD;
	else if (!strcmp(v, "GenuineIntel"))
		c->x86_vendor = X86_VENDOR_INTEL;
	else
		c->x86_vendor = X86_VENDOR_UNKNOWN;
}

struct cpu_model_info {
	int vendor;
	int family;
	char *model_names[16];
};

/* Do some early cpuid on the boot CPU to get some parameter that are
   needed before check_bugs. Everything advanced is in identify_cpu
   below. */
1207
void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218
{
	u32 tfms;

	c->loops_per_jiffy = loops_per_jiffy;
	c->x86_cache_size = -1;
	c->x86_vendor = X86_VENDOR_UNKNOWN;
	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
	c->x86_vendor_id[0] = '\0'; /* Unset */
	c->x86_model_id[0] = '\0';  /* Unset */
	c->x86_clflush_size = 64;
	c->x86_cache_alignment = c->x86_clflush_size;
1219
	c->x86_max_cores = 1;
1220
	c->extended_cpuid_level = 0;
L
Linus Torvalds 已提交
1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
	memset(&c->x86_capability, 0, sizeof c->x86_capability);

	/* Get vendor name */
	cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
	      (unsigned int *)&c->x86_vendor_id[0],
	      (unsigned int *)&c->x86_vendor_id[8],
	      (unsigned int *)&c->x86_vendor_id[4]);
		
	get_cpu_vendor(c);

	/* Initialize the standard set of capabilities */
	/* Note that the vendor-specific code below might override */

	/* Intel-defined flags: level 0x00000001 */
	if (c->cpuid_level >= 0x00000001) {
		__u32 misc;
		cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
		      &c->x86_capability[0]);
		c->x86 = (tfms >> 8) & 0xf;
		c->x86_model = (tfms >> 4) & 0xf;
		c->x86_mask = tfms & 0xf;
1242
		if (c->x86 == 0xf)
L
Linus Torvalds 已提交
1243
			c->x86 += (tfms >> 20) & 0xff;
1244
		if (c->x86 >= 0x6)
L
Linus Torvalds 已提交
1245 1246 1247 1248 1249 1250 1251
			c->x86_model += ((tfms >> 16) & 0xF) << 4;
		if (c->x86_capability[0] & (1<<19)) 
			c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
	} else {
		/* Have CPUID level 0 only - unheard of */
		c->x86 = 4;
	}
1252 1253

#ifdef CONFIG_SMP
1254
	phys_proc_id[smp_processor_id()] = (cpuid_ebx(1) >> 24) & 0xff;
1255
#endif
L
Linus Torvalds 已提交
1256 1257 1258 1259 1260
}

/*
 * This does the hard work of actually picking apart the CPU stuff...
 */
1261
void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1262 1263 1264 1265 1266 1267 1268 1269
{
	int i;
	u32 xlvl;

	early_identify_cpu(c);

	/* AMD-defined flags: level 0x80000001 */
	xlvl = cpuid_eax(0x80000000);
1270
	c->extended_cpuid_level = xlvl;
L
Linus Torvalds 已提交
1271 1272 1273
	if ((xlvl & 0xffff0000) == 0x80000000) {
		if (xlvl >= 0x80000001) {
			c->x86_capability[1] = cpuid_edx(0x80000001);
1274
			c->x86_capability[6] = cpuid_ecx(0x80000001);
L
Linus Torvalds 已提交
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287
		}
		if (xlvl >= 0x80000004)
			get_model_name(c); /* Default name */
	}

	/* Transmeta-defined flags: level 0x80860001 */
	xlvl = cpuid_eax(0x80860000);
	if ((xlvl & 0xffff0000) == 0x80860000) {
		/* Don't set x86_cpuid_level here for now to not confuse. */
		if (xlvl >= 0x80860001)
			c->x86_capability[2] = cpuid_edx(0x80860001);
	}

1288 1289
	c->apicid = phys_pkg_id(0);

L
Linus Torvalds 已提交
1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332
	/*
	 * Vendor-specific initialization.  In this section we
	 * canonicalize the feature flags, meaning if there are
	 * features a certain CPU supports which CPUID doesn't
	 * tell us, CPUID claiming incorrect flags, or other bugs,
	 * we handle them here.
	 *
	 * At the end of this section, c->x86_capability better
	 * indicate the features this CPU genuinely supports!
	 */
	switch (c->x86_vendor) {
	case X86_VENDOR_AMD:
		init_amd(c);
		break;

	case X86_VENDOR_INTEL:
		init_intel(c);
		break;

	case X86_VENDOR_UNKNOWN:
	default:
		display_cacheinfo(c);
		break;
	}

	select_idle_routine(c);
	detect_ht(c); 

	/*
	 * On SMP, boot_cpu_data holds the common feature set between
	 * all CPUs; so make sure that we indicate which features are
	 * common between the CPUs.  The first time this routine gets
	 * executed, c == &boot_cpu_data.
	 */
	if (c != &boot_cpu_data) {
		/* AND the already accumulated flags with these */
		for (i = 0 ; i < NCAPINTS ; i++)
			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
	}

#ifdef CONFIG_X86_MCE
	mcheck_init(c);
#endif
S
Shaohua Li 已提交
1333 1334 1335 1336
	if (c == &boot_cpu_data)
		mtrr_bp_init();
	else
		mtrr_ap_init();
L
Linus Torvalds 已提交
1337
#ifdef CONFIG_NUMA
1338
	numa_add_cpu(smp_processor_id());
L
Linus Torvalds 已提交
1339 1340 1341 1342
#endif
}
 

1343
void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
L
Linus Torvalds 已提交
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377
{
	if (c->x86_model_id[0])
		printk("%s", c->x86_model_id);

	if (c->x86_mask || c->cpuid_level >= 0) 
		printk(" stepping %02x\n", c->x86_mask);
	else
		printk("\n");
}

/*
 *	Get CPU information for use by the procfs.
 */

static int show_cpuinfo(struct seq_file *m, void *v)
{
	struct cpuinfo_x86 *c = v;

	/* 
	 * These flag bits must match the definitions in <asm/cpufeature.h>.
	 * NULL means this bit is undefined or reserved; either way it doesn't
	 * have meaning as far as Linux is concerned.  Note that it's important
	 * to realize there is a difference between this table and CPUID -- if
	 * applications want to get the raw CPUID data, they should access
	 * /dev/cpu/<cpu_nr>/cpuid instead.
	 */
	static char *x86_cap_flags[] = {
		/* Intel-defined */
	        "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
	        "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
	        "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
	        "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL,

		/* AMD-defined */
1378
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
L
Linus Torvalds 已提交
1379 1380
		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL,
1381
		NULL, "fxsr_opt", "rdtscp", NULL, NULL, "lm", "3dnowext", "3dnow",
L
Linus Torvalds 已提交
1382 1383 1384 1385 1386 1387 1388 1389

		/* Transmeta-defined */
		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,

		/* Other (Linux-defined) */
1390
		"cxmmx", NULL, "cyrix_arr", "centaur_mcr", NULL,
1391
		"constant_tsc", NULL, NULL,
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,

		/* Intel-defined (#2) */
1397
		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
L
Linus Torvalds 已提交
1398 1399 1400 1401
		"tm2", NULL, "cid", NULL, NULL, "cx16", "xtpr", NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,

1402 1403 1404 1405 1406 1407
		/* VIA/Cyrix/Centaur-defined */
		NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,

L
Linus Torvalds 已提交
1408
		/* AMD-defined (#2) */
1409
		"lahf_lm", "cmp_legacy", "svm", NULL, "cr8_legacy", NULL, NULL, NULL,
L
Linus Torvalds 已提交
1410 1411
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
1412
		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
L
Linus Torvalds 已提交
1413 1414 1415 1416 1417 1418 1419
	};
	static char *x86_power_flags[] = { 
		"ts",	/* temperature sensor */
		"fid",  /* frequency id control */
		"vid",  /* voltage id control */
		"ttp",  /* thermal trip */
		"tm",
1420 1421
		"stc",
		NULL,
1422
		/* nothing */	/* constant_tsc - moved to flags */
L
Linus Torvalds 已提交
1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447
	};


#ifdef CONFIG_SMP
	if (!cpu_online(c-cpu_data))
		return 0;
#endif

	seq_printf(m,"processor\t: %u\n"
		     "vendor_id\t: %s\n"
		     "cpu family\t: %d\n"
		     "model\t\t: %d\n"
		     "model name\t: %s\n",
		     (unsigned)(c-cpu_data),
		     c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
		     c->x86,
		     (int)c->x86_model,
		     c->x86_model_id[0] ? c->x86_model_id : "unknown");
	
	if (c->x86_mask || c->cpuid_level >= 0)
		seq_printf(m, "stepping\t: %d\n", c->x86_mask);
	else
		seq_printf(m, "stepping\t: unknown\n");
	
	if (cpu_has(c,X86_FEATURE_TSC)) {
1448 1449 1450
		unsigned int freq = cpufreq_quick_get((unsigned)(c-cpu_data));
		if (!freq)
			freq = cpu_khz;
L
Linus Torvalds 已提交
1451
		seq_printf(m, "cpu MHz\t\t: %u.%03u\n",
1452
			     freq / 1000, (freq % 1000));
L
Linus Torvalds 已提交
1453 1454 1455 1456 1457 1458 1459
	}

	/* Cache size */
	if (c->x86_cache_size >= 0) 
		seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size);
	
#ifdef CONFIG_SMP
1460
	if (smp_num_siblings * c->x86_max_cores > 1) {
1461 1462
		int cpu = c - cpu_data;
		seq_printf(m, "physical id\t: %d\n", phys_proc_id[cpu]);
1463
		seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[cpu]));
1464
		seq_printf(m, "core id\t\t: %d\n", cpu_core_id[cpu]);
1465
		seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
1466
	}
L
Linus Torvalds 已提交
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479
#endif	

	seq_printf(m,
	        "fpu\t\t: yes\n"
	        "fpu_exception\t: yes\n"
	        "cpuid level\t: %d\n"
	        "wp\t\t: yes\n"
	        "flags\t\t:",
		   c->cpuid_level);

	{ 
		int i; 
		for ( i = 0 ; i < 32*NCAPINTS ; i++ )
1480
			if (cpu_has(c, i) && x86_cap_flags[i] != NULL)
L
Linus Torvalds 已提交
1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500
				seq_printf(m, " %s", x86_cap_flags[i]);
	}
		
	seq_printf(m, "\nbogomips\t: %lu.%02lu\n",
		   c->loops_per_jiffy/(500000/HZ),
		   (c->loops_per_jiffy/(5000/HZ)) % 100);

	if (c->x86_tlbsize > 0) 
		seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize);
	seq_printf(m, "clflush size\t: %d\n", c->x86_clflush_size);
	seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment);

	seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n", 
		   c->x86_phys_bits, c->x86_virt_bits);

	seq_printf(m, "power management:");
	{
		unsigned i;
		for (i = 0; i < 32; i++) 
			if (c->x86_power & (1 << i)) {
1501 1502 1503 1504 1505
				if (i < ARRAY_SIZE(x86_power_flags) &&
					x86_power_flags[i])
					seq_printf(m, "%s%s",
						x86_power_flags[i][0]?" ":"",
						x86_power_flags[i]);
L
Linus Torvalds 已提交
1506 1507 1508 1509 1510
				else
					seq_printf(m, " [%d]", i);
			}
	}

1511
	seq_printf(m, "\n\n");
L
Linus Torvalds 已提交
1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536

	return 0;
}

static void *c_start(struct seq_file *m, loff_t *pos)
{
	return *pos < NR_CPUS ? cpu_data + *pos : NULL;
}

static void *c_next(struct seq_file *m, void *v, loff_t *pos)
{
	++*pos;
	return c_start(m, pos);
}

static void c_stop(struct seq_file *m, void *v)
{
}

struct seq_operations cpuinfo_op = {
	.start =c_start,
	.next =	c_next,
	.stop =	c_stop,
	.show =	show_cpuinfo,
};
1537

1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556
#ifdef CONFIG_INPUT_PCSPKR
#include <linux/platform_device.h>
static __init int add_pcspkr(void)
{
	struct platform_device *pd;
	int ret;

	pd = platform_device_alloc("pcspkr", -1);
	if (!pd)
		return -ENOMEM;

	ret = platform_device_add(pd);
	if (ret)
		platform_device_put(pd);

	return ret;
}
device_initcall(add_pcspkr);
#endif